SplitIntoTrainTest: Split data into training and testing

In kaylafrisoli/ActiveRL:

Description

Split a dataset with unique identifiers into training and testing datasets with a specified probability

Usage

SplitIntoTrainTest(RLdata, unique.ids, seed = NULL, prob.of.train = NULL)

Arguments

RLdata	a data frame containing the records to be matched
unique.ids	a vector containing the true unique identifiers of the records in RLdata. It should be of length nrow(RLdata)
seed	an integer specifying the seed to set before randomly dividing RLdata
prob.of.train	a probability between 0 and 1 that a given record is in the training dataset

Value

seed	the seed used
prob.of.train	the probability a record is in the training dataset
train	a vector of TRUE/FALSE values corresponding to whether the record is in the training dataset
test	a vector of TRUE/FALSE values corresponding to whether the record is in the testing dataset
training.data	the training dataset
testing.data	the testing dataset
train.ids	unique ids for the training dataset
test.ids	unique ids for the testing dataset

kaylafrisoli/ActiveRL documentation built on May 20, 2019, 7:53 a.m.